A central goal of large scale neuroscience initiatives is to provide a functional and structural description of the canonical cortical column that will generalize across columns, cortical regions, and species. One of the most studied columns is in mammalian visual cortex, where the thalamocortical (TC) input projects to layer 4 of cortex. Prior neuroanatomical studies suggest the number of the TC inputs is low (<10% of total input) - hence weak and sparse - which has led to theories requiring strong recurrent amplification from intracortical (IC) afferents to achieve observed response levels. Based on refined immunohistochemical processing and innovative serial electronmicroscopic techniques, our preliminary results indicate that the weak and sparse theories need substantial revision. We find that TC input is not as sparse as previously estimated, but more importantly the 3D reconstruction of TC terminals shows they have a volume 4 -10 times greater than IC terminals. In the current proposal we will use new techniques to characterize the major synaptic inputs in two crucial gateways in the cortical microcircuit, layers 4 and 2/3. In Aim 1 we will measure the number and 3D size of TC and IC terminals in layer 4, as well as the number of vesicles and size of regions of postysynaptic density. The expected outcome is that individual TC efficacy based on size and vesicle number will be greater than individual IC efficacy. This suggests that the population of TC terminals onto layer 4 neurons will be equally effective as the larger population of smaller IC terminals: a result with major implications for models using large scale recurrent amplification. In Aim 2 we will compare the number, size and vesicle density of terminals arising from feed-forward (FF) input from layer 4 to layer 2/3, and from feed-back (FB) terminals arising from intra-cortical lateral feedback (IC FB) or interareal (IA FB) feedback to neurons in layer 2/3. The different populations will be revealed by microinjection of anterograde tracer into layer 2/3 or 4 of area 17, and extrastriate areas 18 or Ssy. The expected outcome is that FF inputs will be larger and more numerous than either FB population: providing the synaptic weight assignment within the canonical microcircuit. Overall our goal is to establish a set of techniques (immunohistochemistry and FIB/SEM) that will enable quantitative comparisons among areas and species of the fine structure of cortical circuits.